Ex-Scale AI general manager who joined Handshake AI in June 2025 to own strategy, positioning and the innovation roadmap. He led the Cleanlab acquisition narrative and is Handshake AI's clearest strategic voice on expert quality.
high confidenceSan FranciscoAt Handshake AI since 2025Updated 2026-09-19
Background
At Scale AI he was General Manager and Director of Product & AI for the International/Global Public Sector division, with earlier roles in quality infrastructure and strategic product management. Before that: Bain & Company consultant, international expansion lead at a consumer products company (The Org lists BARK), and a startup founder. BA in Pure Mathematics and MA in Economics from Boston University.
What they run now
Long-term strategy, market positioning and innovation roadmap; data quality via the Cleanlab team.
Career
2025-06 – presentChief Strategy & Innovation Officer, Handshake AI
2025GM & Director of Product and AI, International Public Sector, Scale AI
—Consultant, Bain & Company
—Boston University, BA Pure Mathematics; MA Economics
On the record
Supply advantage · 2025
Argues Handshake's verified student/PhD network lets it find specific experts in minutes where rivals take weeks. source
Experts undervalued · 2025
Says no company in the industry has fully internalised how critical experts are, and that post-training needs strong writing, attention to detail and real domain expertise. source
Data quality · 2026-01
Framed the Cleanlab acqui-hire as buying years of focus on data-quality assessment. source
Recent posts
38 posts archived · most engaged first, then the latest
AI data factories like Handshake, Scale, Surge, and Mercor have a big role to play in the advancement of AI at the frontier and in the enterprise. This is my full thesis. x.com/i/article/2056792438342385…
There is an astounding amount of human <> agent collaboration required to produce great RL environments and tasks.
The skill of authoring this kind of data will increasingly become a critical full-time, high-skilled job, both at data companies and within enterprises.
It's become increasingly clear that most people dramatically underestimate how essential evals for every enterprise workflow within every individual enterprise are going to be
Running long-horizon, agentic data projects is the best preparation for building robust enterprise AI apps bc it forces you to build thoughtful expert <> agent interactions and optimize LLM usage. Data companies have a massive advantage in serving the enterprise.
We now spend more on LLMs than on headcount. That does not make human operators, and especially experts, less valuable. It actually makes them more productive and more valuable. Purely synthetic data without experts embedded in the authoring process doesn't work, and data companies that don't dig into their own data will ship slop. They won't produce the nuance, realism, or complexity the task demands, and the labs struggle to catch all these issues because they rely on agentic graders that have the same blind spots.
But the token spend is significant, and it goes to three places: (1) hybrid synthetic + expert environment / task authoring, (2) rollouts to validate task complexity, and (3) QC on the data. You cannot produce a high-quality long-horizon agentic task without LLMs across all three buckets. We also use agents to automate internal processes, but that is a small fraction of the bill. Any data company telling you the bulk of its LLM spend is internal automation is lying.
Producing this data at high quality and reasonable cost comes down to two problems.
(1) the expert-agent interaction paradigm - how the expert creates, reviews, interprets, and audits the LLM's outputs (synthetic input files, prompts, rubrics, and so on). Solving this means knowing precisely where agents fail or underperform, designing the interface through which experts validate those outputs, and building the tools that let experts steer the agent going forward.
(2) optimizing every QC check for the right tradeoff between cost, quality, and latency. That means constructing complex and diverse eval sets, configuring an agentic system with the right models (often chained prompts, guardrail LLM calls, and deterministic checks), running the eval, then iterating on the harness, prompts, chaining, and model choice until the mix is optimal.
Both are now core skills of our SPLs.
Enterprise deployments follow the exact same pattern. Understand the workflow and where models fail or underperform, design the user-agent interaction to mitigate those failures, build evals that test the agentic system across the workflow, and optimize the system against those evals. Running data pipelines forces us through this loop hundreds of times a month across a wide range of project types, and we have to build product and infrastructure to do it fast and repeatably.
You cannot build an optimal enterprise deployment without these muscles, and the largest data companies have already done it hundreds of times.
Knowing whether agents actually work on economically valuable tasks in environments as messy as production is a core bottleneck for AI.
Launching ATLAS-Finance, the most realistic finance eval to date. The best model succeeds on 12% of tasks. Proud of the team!
Curtis Northcutt — Cleanlab CEO, acqui-hired Jan 2026
Why it matters here
The person who decides which new data verticals Handshake AI enters. A taste/creative vertical would be a strategy-and-innovation call. High relevance.
How to reach
Active on X. Frame a pitch around a quality problem (measuring taste reliability) rather than supply, since he cares about quality differentiation (speculation).